Slide 1: "Measures of Association"

1. Two variables have a strong statistical relationship with one another if they appear to move
together:

So, what this sentence is saying is that if two variables always move together – both going up or
both going down – they have a strong statistical relationship. It's like they're buddies, always
sticking together.

2. When two variables appear to be related, you might suspect a cause-and-effect relationship:

o For example, imagine that whenever it's hot outside, people buy more ice cream. You
might think, "Ah, it’s hot, so people are buying ice cream to cool down." Here, you
suspect that the heat is causing more ice cream sales — that’s a cause-and-effect
relationship.

3. Sometimes, however, statistical relationships exist even though a change in one variable is not
caused by a change in the other:

o But, let’s not jump to conclusions! Sometimes, two things happen together by chance.
For instance, you might notice that people who eat according to eatclean diet also
tend to read more books. Right?

o It doesn’t necessarily mean eating healthily causes people to read more — there
could be some other reason, like these people just generally have healthier habits!

Slide 2: "Measures of Association: Covariance" (Hiệp ph/ương sai)

1. Covariance:

o Covariance helps us measure how two things change together. For example,
covariance tells us if you and your friend’s spending on coffe each month go up and
down together or if they go in opposite directions.

When analyzing covariance, we look at the number calculated from separate formulas for the
population and the sample.

2. Population Covariance: (Hiệp phương sai tổng thể)

o When we calculate covariance for an entire group or population (like all the people in
a city), we use a specific formula. This is like figuring out if all students in a school
perform better in math when they study more hours. You’d use the formula to find out
if there’s a general trend across everyone in the group. In Excel, the formula
=COVARIANCE.P(array1, array2) can help do this quickly.

3. Sample Covariance: (Hiệp phương sai của mẫu)

o But, if you’re only looking at a smaller group or a sample, like a class of 30 students
instead of the whole school, the formula changes a bit. It’s like taking a sneak peek at
a few people to guess what the bigger group might be doing. For this, you would use
=COVARIANCE.S(array1, array2) in Excel.
 4. Explanation of Covariance:

o So, in simple terms, covariance is like asking: "Do two things move in the same
direction or not?" If they do, the covariance will be positive (+). If one goes up while
the other goes down, it’ll be negative (-). It’s a way to see if there's some kind of
pattern in how things change together.

Slide 3: "Example 4.20: Computing the Covariance"

1. Example Using Colleges and Universities Data:

o Here, we have a practical example includes0 data from different colleges about their
graduation rates and the median SAT scores of their students. We want to find out if
schools with higher SAT scores tend to have higher graduation rates too. If they do,
this could mean there’s some kind of relationship between SAT scores and how many
students graduate.

2. Data Table Explanation:

o In the table, we have a lot of numbers. But don’t worry! Each column is helping us
calculate how much each school's graduation rate and SAT score differ from the
average. We then multiply those differences together to see if there’s a pattern — do
schools with above-average SAT scores also have above-average graduation rates?

3. Covariance Calculation:

o Finally, we use these calculations to get a number, the covariance, using Excel’s
=COVARIANCE.S function. The result here is 263.37. A positive number means,
generally, when SAT scores go up, graduation rates go up too, suggesting they are
related.

Slide 4:

Imagine we're looking at two things, like ice cream sales and temperature. Covariance is a measure of
how these two things move together.

 The first part says that if the absolute value of covariance is big, it means there's a strong
connection between the two things. In our ice cream example, a high covariance would suggest
that as the temperature goes up, ice cream sales also go up.

 The second part is about the sign. If the covariance is positive, it means that when one thing
increases, the other thing also increases. This is a direct relationship, like ice cream sales and
temperature. But if the covariance is negative, it means that when one thing increases, the other
thing decreases. This is an inverse relationship, like the amount of time you spend studying and
the number of mistakes you make on a test.